Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

Dataset: 80% C4 (Web Text) and 20% Python Code

Neuron 5331 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 2.1340. Min Range: -2.1340

Max Act: 2.1340. Min Act: -0.0000

Data Index: 1549400 (Python Code)

Max Activating Token Index: 812

Text #1

Max Range: 2.1340. Min Range: -2.1340

Max Act: 1.9203. Min Act: 0.0001

Data Index: 1102232 (C4 (Web Text))

Max Activating Token Index: 26

Text #2

Max Range: 2.1340. Min Range: -2.1340

Max Act: 1.8937. Min Act: -0.0000

Data Index: 1549134 (Python Code)

Max Activating Token Index: 902

Text #3

Max Range: 2.1340. Min Range: -2.1340

Max Act: 1.8635. Min Act: 0.0001

Data Index: 332921 (C4 (Web Text))

Max Activating Token Index: 8

Text #4

Max Range: 2.1340. Min Range: -2.1340

Max Act: 1.7411. Min Act: 0.0001

Data Index: 1148016 (C4 (Web Text))

Max Activating Token Index: 10

Text #5

Max Range: 2.1340. Min Range: -2.1340

Max Act: 1.6319. Min Act: -0.0000

Data Index: 361428 (C4 (Web Text))

Max Activating Token Index: 77

Text #6

Max Range: 2.1340. Min Range: -2.1340

Max Act: 1.6679. Min Act: 0.0000

Data Index: 981203 (C4 (Web Text))

Max Activating Token Index: 6

Text #7

Max Range: 2.1340. Min Range: -2.1340

Max Act: 1.6196. Min Act: 0.0001

Data Index: 1220741 (C4 (Web Text))

Max Activating Token Index: 18

Text #8

Max Range: 2.1340. Min Range: -2.1340

Max Act: 1.4822. Min Act: -0.0000

Data Index: 1618147 (Python Code)

Max Activating Token Index: 844

Text #9

Max Range: 2.1340. Min Range: -2.1340

Max Act: 1.5473. Min Act: 0.0000

Data Index: 897969 (C4 (Web Text))

Max Activating Token Index: 186

Text #10

Max Range: 2.1340. Min Range: -2.1340

Max Act: 1.4577. Min Act: 0.0001

Data Index: 1240874 (C4 (Web Text))

Max Activating Token Index: 65

Text #11

Max Range: 2.1340. Min Range: -2.1340

Max Act: 1.4777. Min Act: 0.0001

Data Index: 535 (C4 (Web Text))

Max Activating Token Index: 108

Text #12

Max Range: 2.1340. Min Range: -2.1340

Max Act: 1.4414. Min Act: 0.0000

Data Index: 1217847 (C4 (Web Text))

Max Activating Token Index: 365

Text #13

Max Range: 2.1340. Min Range: -2.1340

Max Act: 1.4288. Min Act: -0.0000

Data Index: 873733 (C4 (Web Text))

Max Activating Token Index: 67

Text #14

Max Range: 2.1340. Min Range: -2.1340

Max Act: 1.5051. Min Act: 0.0001

Data Index: 866533 (C4 (Web Text))

Max Activating Token Index: 16

Text #15

Max Range: 2.1340. Min Range: -2.1340

Max Act: 1.4916. Min Act: -0.0000

Data Index: 1540192 (Python Code)

Max Activating Token Index: 527

Text #16

Max Range: 2.1340. Min Range: -2.1340

Max Act: 1.4898. Min Act: -0.0000

Data Index: 1386126 (Python Code)

Max Activating Token Index: 621

Text #17

Max Range: 2.1340. Min Range: -2.1340

Max Act: 1.5508. Min Act: 0.0001

Data Index: 742457 (C4 (Web Text))

Max Activating Token Index: 23

Text #18

Max Range: 2.1340. Min Range: -2.1340

Max Act: 1.5084. Min Act: 0.0001

Data Index: 145493 (C4 (Web Text))

Max Activating Token Index: 31

Text #19

Max Range: 2.1340. Min Range: -2.1340

Max Act: 1.5354. Min Act: -0.0000

Data Index: 1559238 (Python Code)

Max Activating Token Index: 815